Message-ID: <002001c1ee43$25ea0dc0$743f01d5@default>
From: "Alan Melia" <Alan.Melia@btinternet.com>
To: "LF-Group" <rsgb_lf_group@blacksheep.org>
Subject: LF: Mains Harmonics ??
Date: Sun, 28 Apr 2002 00:25:01 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Hi All, here is one to shoot at !! If as Cesare says the interference he
gets are straight harmonics of the 50Hz mains and they are the 2750th
harmonic, should they not be 1/ (2750)^2 down (-72dB) on the fundamental ??
which ought to be quite small in the cables so even smaller radiated.

 I see drifty lines with 50Hz and 100Hz spacing, but a quick check at lower
frequencies with teh Rx or SA shows that these are mains modulated signals
with a funamental at 30 to 40kHz and emanate mainly from switch-mode power
supplies, PCs and TVs being the worst offenders. If they are such, the drift
is due to the switcher oscillator not the mains so you cannot lock a comb to
it. If it genuinely is straight mains frequency harmonics then it should be
possible to lock a switched capacitor comb filter to this.

Most SMPSU QRM grids have a spacing of twice the mains frequency as there is
a bridge rectifer involved in the primary side of the PSU, though often the
intermediate 50Hz is seen weakly. Local TVs seem to radiate the 40kHz via
the braid of the aerial coax, so a standard high-pass TVI filter with a
braid-breaker should work on reverse (Mike G3XDV has previously reported
this solution). If I could actually get some out of the RSGB I might be able
to confirm this.

I was able to make quite a good reduction in the locally generated lines by
ensuring that cables were tidied up and that no aerial coax ( for any
frequency ) ran near any mains cable. Wall-wart supplies to feed the RX have
the 12v cable wrapped into a coil and taped round a ferrite rod. PC suppies
were opened and fitted with screened IEC filtered sockets (the 3amp version
because it has the biggest inductors) This later was the most dramatic as I
often have 4 PCs running in the house. All feeds to soundcards are isolated
via audio transformers. All my noise now is from neighbour's TVs.  I can
tell which channel some are watching by the cadence of the FM inducted on
the PSU by the audio load, particularly at news time (why do they all switch
channel when those interesting Party Political broadcasts come on ?? could I
use blackmail ??....."put this filter in or I'll tell Mr Blair you switched
him off!" ).

Another gotcha is that loops dont always help, they will couple magetically
to TV and monitor scan coils at quite large distances (10m+) , the givaway
...TVs scan at mains and PC monitors at 70Hz or about 90Hz, both are locked
and are "crystal" stable lines. Also the direction is at right-angles to the
normal loop maximum direction. Fortunately most of the hamonics of the line
scans are outside the band.

Cheers de ALan G3NYK
alan.melia@btinternet.com